Effects of σ Phase on the SCC Susceptibility and the Hydrogen Embrittlement Susceptibility of Super Duplex Stainless Steel F55

Abstract

While duplex stainless steel has excellent mechanical properties and corrosion resistance, intermetallic compounds called σ phase easily precipitate. In this study, SSRT was performed using specimens with different amounts of σ phase precipitation. In corrosive solution, specimens with precipitated σ phase tended to have a more pronounced decrease in ductility. Moreover, SSRT was performed in a similar corrosive solution while continuously charging hydrogen by cathodic charging method. As a result, there was no correlation between the amounts of σ phase precipitation and mechanical properties. Therefore, the σ phase is considered to promote SCC while not promoting hydrogen embrittlement.

Fig. 13 Rate of cracks passing through σ phase on the surface of super duplex stainless steel F55 after SSRT in air and under a constant potential of −0.5 V (vs. Ag/AgCl). (online color)

1. Introduction

Duplex stainless steel is composed of ferritic phase and austenitic phase. It has excellent mechanical properties and resistance to pitting corrosion and Stress Corrosion Cracking (SCC) in chloride environments. Moreover, because the Ni content is smaller than that of austenitic stainless steel, application of duplex stainless steel is extended in chloride environments such as chemical plants or seawater and in severe corrosive environments such as oil or gas fields [1–5]. However, due to the reduction of oil reserves in recent years, it has become necessary to redevelop oil or gas fields which had been difficult to develop until now. Therefore, the demand for parts of oil drilling equipment shifted from castings to forgings with superior strength. From this background, a steel forging that meets the ASTM (America Society for Testing and Materials)-F55 standard was prepared. The material of these steel forgings is the super duplex stainless steel F55 used in this study. This material is a relatively new type of stainless steel and is attracting attention for its characteristics of both strength and corrosion resistance.

However, it was reported that σ phase is easily precipitated in the duplex stainless steel during heating at around 1173 K. The σ phase is a hard and brittle intermetallic compound composed mainly of Fe, Cr and Mo. It is known that the precipitation of σ phase causes deterioration of ductility and corrosion resistance of the material [6–8]. Super duplex stainless steel has higher corrosion resistance in chloride environments by increasing the Cr, Mo, and N contents. However, it is thought that the σ phase precipitates are easily formed because it contains a large amount of Cr and Mo which are the constituent elements of the σ phase. If it were precipitated, its effect would be concerned. Nagae et al. investigated the effect of σ phase precipitated in super duplex stainless steel on corrosion resistance. They reported that when σ phase precipitates, denuded zones of Cr and Mo are formed around the σ phase, and the passivation film in the area is preferentially dissolved, resulting in a decrease in corrosion resistance [9]. Active Path Corrosion-SCC (APC-SCC) is a phenomenon in which crack propagation starts with corrosion or pitting corrosion of the slip line, and the crack propagates by preferential dissolution of the area with reduced corrosion resistance. In the case of specimens with σ phase precipitation, the Cr and Mo denuded zones around the σ phase, which are active regions, act as crack propagation paths and may significantly cause APC-SCC.

It has been reported in many papers that a phenomenon called hydrogen embrittlement occurs remarkably in duplex stainless steel. This is a phenomenon in which hydrogen penetrates into the material and makes it brittle [10–12]. Hydrogen is thought to be trapped in lattice defects such as grain boundaries, dislocations, vacancies and precipitates in metallic materials. Differences in the microstructure and crystal structure of metallic materials are said to significantly affect the hydrogen embrittlement properties [10, 13, 14]. Slow Strain Rate Test (SSRT) is often used to evaluate hydrogen embrittlement properties. It is known that there is a correlation between the decrease in ductility and the strain rate due to hydrogen embrittlement. In general, the smaller the strain rate, the smaller the maximum strain due to hydrogen embrittlement [15, 16]. When plastic deformation takes place slowly as in the case of SSRT, diffusion of hydrogen can follow up dislocation movements and hydrogen embrittlement susceptibility become higher due to the strong interaction between dislocations and hydrogen atoms. On the other hand, when plastic deformation takes place rapidly, dislocations move rapidly and hydrogen atoms cannot catch up dislocations. As interaction between them becomes weak, it is assumed that hydrogen embrittlement susceptibility becomes lower [17]. Typical methods for introducing hydrogen into a specimen are a method of exposing the specimen to high-pressure hydrogen gas and a cathodic charging method. Deterioration of mechanical properties due to hydrogen embrittlement has been reported in both methods [4, 18].

Although many studies have been conducted on the σ phase, SCC and hydrogen embrittlement of stainless steel, there are few research reports that focus on all of these under the same environment [1–20]. In this study, to clarify the effects of σ phase on the SCC susceptibility and the hydrogen embrittlement susceptibility of super duplex stainless steel F55, five types of specimens with different precipitation amounts of σ phase were prepared. The effect of SCC on the mechanical properties of each specimen was investigated by performing SSRT in a mixed aqueous solution of sulfuric acid and sodium chloride, which is a harsh environment for stainless steels. Moreover, the effect of hydrogen embrittlement on the mechanical properties of each specimen was investigated by performing SSRT while hydrogen was continuously introduced using the cathodic charging method. The similar solution was used for SSRT in a corrosive solution and under cathodic charging. After SSRT, the fracture surface was observed in order to clarify the relationship between the test environment, metallographic structure and the fracture morphology. In addition, surface cracks were observed to reveal the susceptibility to cracking of each phase and interface due to SCC and hydrogen embrittlement.

2. Experimental Procedure

2.1 Specimens

Newly developed F55 super duplex stainless steel was prepared and used as the specimen. Table 1 shows its chemical composition. It meets the ASTM-F55 standard and has a relatively high Cr content of 25.0 mass% compared to other stainless steels. In addition, it contains Ni, Mo, W, N and others. Pitting Resistance Equivalent (PRE = mass% Cr + 3.3 × mass% Mo + 0.5 × mass% W + 16 × mass% N) is an index indicating corrosion resistance of stainless steel. Super duplex stainless steel is defined as a one whose PRE is between 40 and 45 [21]. It can be confirmed that the F55 used in this study is a super duplex stainless steel with a PRE of 42. In this study, five types of specimens with different amounts of σ phase precipitation were prepared to clarify the effect of sigma phase on the mechanical properties of F55 under SCC and hydrogen embrittlement environments. The amounts of σ phase precipitation can be controlled by changing heat treatment temperature and time. The heat treatment time was 5.4 ks and then water quenched. The heat treatment temperatures were 1173 K corresponding to easy precipitation zone and additional 1223 K, 1248 K, 1273 K and 1373 K [6–8]. Figure 1 shows the surface microstructure of F55 after heat treatment at each temperature. Brown phases indicated by α are ferritic phases, white phases indicated by γ are austenitic phases and black granular phases shown within the ferritic phases are σ phases. Moreover, a white contrast is seen in the ferritic phase. This indicates that part of the ferritic phase changed to the austenitic phase during the heat treatment process. In this paper, this is called the secondary austenitic phase. No σ phase precipitate was observed in the specimen treated at 1373 K. The σ phase precipitates were observed at the grain boundary between a ferritic phase and an austenitic phase in the specimen treated at 1273 K. The σ phase precipitated within the ferritic phase as well as in the grain boundary in a specimen treated at 1248 K, 1223 K and 1173 K. The area ratio of σ phase of these heat-treated specimens estimated using image analysis software. Table 2 shows the estimated results. The lower the heat treatment temperature, the larger the area ratio of σ phase in the five types of specimens used in this study.

Table 1 Chemical composition of super stainless steels F55, ferritic stainless steel SUS445J1 and austenitic stainless steel SUSF310 (mass%).

Fig. 1

Optical micrographs of super duplex stainless steel F55 after heat treatment at (a) 1373 K, (b) 1273 K, (c) 1248 K, (d) 1223 K and (e) 1173 K. (online color)

Table 2 Area ratio of σ phase after heat treatment at 1373 K, 1273 K, 1248 K, 1223 K and 1173 K.

Ferritic stainless steel SUS445J1 and austenitic stainless steel SUSF310 were prepared in order to clarify the hydrogen embrittlement susceptibility of each phase in duplex stainless steel. Table 1 shows the chemical composition of each.

2.2 Polarization curve measurements

Polarization curve measurements were performed to investigate the corrosion behavior of each specimen. Test solution was used a mixed aqueous solution of 2.5 mol·dm⁻³ H₂SO₄ + 0.2 mol·dm⁻³ NaCl, which stainless steel is susceptible to SCC [22].

It was prepared precisely at room temperature adjusted to 298 K using an electronic balance and a hydrometer. The test was performed by keeping a H-shaped glass cell filled with a test solution which residual oxygen was purged using pure nitrogen gas at 298 K in a constant temperature bath. Each specimen having a different area ratio of σ phase was wet polished until glossy surface was revealed using water resistant emery papers with grit sizes of 80 to 2000 in sequential order and 300 s ultrasound cleaning in an acetone solution was applied to obtain a final specimen. Specimen coated with a silicon rubber leaving an area of 1.0 × 10⁻⁴ m² as an activation area was used as a working electrode and a Pt electrode with an area of 1.2 × 10⁻³ m² was used as the counter electrode. A working electrode measured potential of the surface boundary by a Luggin capillary attached to the vicinity of an exposed surface and was collated to the referential electrode via double junction being connected to a potentiostat [Bio-Logic Instruments Ltd SP-150] controlled by a computer for potential sweeping. It should be noted that potentials described in this paper were referred to an Ag/AgCl (3.33 mol·dm⁻³ KCl) electrode. Electrode potential of the Ag/AgCl is +0.206 V at 298 K with respect to a standard hydrogen electrode (SHE). Potential was swept from −0.7 V (vs. Ag/AgCl) to +1.0 V (vs. Ag/AgCl) at a sweep rate of 0.5 mV·s⁻¹. Surface boundary potential and logarithms of current density were recorded.

2.3 SSRT

Specimen size was small enough not to cause significant difference in cooling rate between bulk surface and inside of the specimen, a large difference in the precipitation amount of the σ phase was avoided. A specimen used for tensile test with a gage section of 20 mm in length, 4 mm in width and 2 mm in thickness was prepared. Specimens for tensile tests were wet polished until glossy surfaces were revealed using water resistant emery papers with grit sizes of 80 to 2000 in sequential order and 300 s ultrasound cleaning in an acetone solution was applied.

SSRT was performed in air, in corrosive solution, and under cathodic charging. First, SSRT was performed in air to investigate the basic mechanical properties of each specimen. In corrosive solution, a mixed aqueous solution of 2.5 mol·dm⁻³ H₂SO₄ + 0.2 mol·dm⁻³ NaCl was used to investigate the mechanical properties in an environment where SCC is highly likely to occur. Under cathodic charging, SSRT was performed using a mixed aqueous solution of 2.5 mol·dm⁻³ H₂SO₄ + 0.2 mol·dm⁻³ NaCl and maintaining the potential in the cathodic region. The set potential of SSRT under cathodic charging was determined based on the results of polarization curve measurements, and was set to a potential that showed a large current density in the cathodic region and had small variations depending on the specimen. Mechanical properties under hydrogen embrittlement were investigated by continuously charging each specimen with hydrogen. A specimen was coated with a silicon rubber except a gage section to prevent influence of the solution on outsides of a gage section and used as a working electrode. A Pt electrode was used as the counter electrode. The specimen was collated to the referential electrode of Ag/AgCl (3.33 mol·dm⁻³ KCl) via double junction and was connected to a potentiostat to regulate the potential. In this study, hydrogen charging was not performed in advance and cathodic charging was started at the same time as the SSRT. The strain rate was set at 1.25·10⁻⁶ s⁻¹ under all test environments.

3. Results and Discussion

3.1 Polarization curve measurements

Figure 2 shows results of polarization curve measurements for super duplex stainless steel F55 in a 2.5 mol·dm⁻³ H₂SO₄ + 0.2 mol·dm⁻³ NaCl solution. All polarization curves showed typical polarization behavior such as active and inactive state regions characteristic of stainless steel. Moreover, two active state peaks of ferritic phase and austenitic phase characteristic to duplex stainless steel can be confirmed. Once a passivation film was formed, each specimen showed a similar curve. However, there were differences in current density among specimens in an active state region between a corrosion potential E_corr and an inactive state potential around −0.15 V (vs. Ag/AgCl). Larger the amount of σ phase, current density increased slightly. The anodic reaction is enhanced as the area ratio of σ phase increases, and the presence of the σ phase inhibits the formation of the passivation film. Therefore, it is considered that the formation of the passivation film is delayed. On the other hand, once a passivation film was formed, it was confirmed that the corrosion resistance was equivalent regardless of the area ratio of the σ phase. The set potential of SSRT under cathodic charging was determined based on the results of polarization curve measurements. To investigate mechanical properties under the hydrogen embrittlement environment in this study, a potential of −0.5 V (vs. Ag/AgCl) which shows a large current density in the cathodic region in all polarization curves was applied only to the gage section of the specimen and SSRT was performed.

Fig. 2

Polarization curve of Super Duplex stainless steel F55. (online color)

Figure 3 shows results of anodic polarization curve measurements for ferritic stainless steel SUS445J1 and austenitic stainless steel SUSF310 in a 2.5 mol·dm⁻³ H₂SO₄ + 0.2 mol·dm⁻³ NaCl solution. A potential of −0.5 V (vs. Ag/AgCl) which shows a large current density in the cathodic region in all polarization curves was applied only to the gage section of the specimen and SSRT was performed.

Fig. 3

Polarization curve of ferritic stainless steel SUS445J1 and austenitic stainless steel SUSF310. (online color)

3.2 SSRT

3.2.1 Super duplex stainless steel

Figure 4(a)–(c) show the stress-strain curves of five types of super duplex stainless steel F55 with different area ratio of σ phase under each test environment. Table 3 shows the average values of the ultimate tensile strength and maximum strain obtained by repeating the same experiment multiple times. The ultimate tensile strength and maximum strain decrease along with the increase in area ratio of σ phase in air and in corrosive solution. It is considered that the mechanical properties of the material decrease because the σ phase which is hard and brittle precipitates at the interface between the ferritic phase and austenitic phase and within the ferritic phase grains [6]. The ultimate tensile strength and maximum strain of SSRT in corrosive solution are lower than those in air because of the severe environment where SCC is likely to occur. Under a constant potential of −0.5 V (vs. Ag/AgCl), the ultimate tensile strength and maximum strain showed constant low values regardless of the area ratio of σ phase. In the cathodic region, the influence of SCC is reduced because the anodic reaction is suppressed, but the mechanical properties were lower than in corrosive solution. This suggests the high hydrogen embrittlement susceptibility of super duplex stainless steel F55. Okayasu et al. reported that the interface between the ferritic phase and the austenitic phase in duplex stainless steel has a high internal strain, which acts as an effective hydrogen trap site and causes hydrogen embrittlement [10]. Figure 5(a), (b) show an influence of the area ratio of σ phase on the ultimate tensile strength and maximum strain of each specimen. Comparing the ultimate tensile strength in corrosive solution with that in air, a reduction of 15.1%, 16.7%, 7.7%, 13.2% and 10.9% was confirmed at the area ratio of σ phase of 0%, 8.7%, 14.4%, 20.9% and 35.2% respectively. There was no correlation between the ultimate tensile strength and area ratio of σ phase. Comparing the maximum strain in corrosive solution with that in air, a reduction of 35.7%, 53.2%, 48.7%, 41.6% and 20.8% was confirmed at the area ratio of σ phase of 0%, 8.7%, 14.4%, 20.9% and 35.2% respectively. With the exception of the area ratio of σ phase of 35.2%, specimens with precipitated σ phase showed a larger reduction rate of maximum strain than specimens without precipitated σ phase. This suggests that the presence of σ phase promotes SCC. It is known that there is a Cr and Mo depleted zone around the σ phase, and this region is preferentially dissolved [9]. This is thought to have increased APC-SCC susceptibility and decreased ductility. In the case of area ratio of σ phase of 35.2%, it is thought that the influence of SCC was smaller than σ phase embrittlement, and no significant decrease in ductility occurred.

Fig. 4

Stress-Strain curves of super duplex stainless steel F55 by SSRT under each test environment at 298 K. (online color)

Table 3 Ultimate tensile strength and maximum strain of super duplex stainless steel F55 derived from Stress-Strain curves.

Fig. 5

The effect of area ratio of σ phase (%) on (a) ultimate tensile strength (UTS), (b) maximum strain under the SSRT. (online color)

Comparing the ultimate tensile strength under a constant potential of −0.5 V (vs. Ag/AgCl) with that in air, a reduction of 35.8%, 26.9%, 12.6%, 13.2% and 16.6% was confirmed at the area ratio of σ phase of 0%, 8.7%, 14.4%, 20.9% and 35.2% respectively. Moreover, comparing the maximum strain under a constant potential of −0.5 V (vs. Ag/AgCl) with that in air, a reduction of 89.1%, 78.1%, 57.3%, 46.8%, 33.3% was confirmed at the area ratio of σ phase of 0%, 8.7%, 14.4%, 20.9% and 35.2% respectively. As the area fraction of σ phase increases, the rate of decrease in both ultimate tensile strength and maximum strain decreases. Therefore, the presence of σ phase does not significantly promote hydrogen embrittlement.

It is difficult to determine just from the SSRT results whether σ phase embrittlement or hydrogen embrittlement dominates the embrittlement behavior of the specimens with σ phase precipitation. This is discussed based on the results of fracture surface observation in Section 3.3 and surface observation in Section 3.4. Nakade et al. prepared the specimens with a pseudo-duplex structure of austenitic phase and σ phase by heating a duplex stainless steel for a long time. The presence of the σ phase is considered to increase the hydrogen embrittlement susceptibility as a result of tensile tests after cathodic charging to these specimens [23]. However, this study does not support this result. In the case of Nakade’s study, specimens with a pseudo-duplex structure of austenitic phase and σ phase was prepared by heat treatment at 1023 K. During the heat treatment process, most of the ferritic phase transforms to a secondary austenitic phase. Therefore, it is considered that there is a large residual stress in the regions where the original ferritic phase existed. The residual stresses are created by the phase transformation could enhance the susceptibility to hydrogen embrittlement [11].

3.2.2 Ferritic stainless steel and austenitic stainless steel

As a result of the SSRT under a hydrogen embrittlement environment using the super duplex stainless steel F55 with area ratio of σ phase of 0% described in Section 3.2.1, it was confirmed that there was a decrease in ductility which is thought to be caused by hydrogen embrittlement. However, it is not clear that Which of ferritic phase and austenitic phase caused decrease in ductility. In Section 3.2.2, the same SSRT as in Section 3.2.1 was performed using ferritic stainless steel SUS445J1 and austenitic stainless steel SUSF310, and the hydrogen embrittlement susceptibility of each phase in duplex stainless steel was clarified. Figure 6(a) shows the stress-strain curves of ferritic stainless steel SUS445J1 under each test environment. Table 4 shows the ultimate tensile strength and maximum strain obtained from the stress-strain curves. Under a constant potential of −0.5 V (vs. Ag/AgCl), the maximum strain is the smallest and it is 70.3% less than in air. It can be confirmed that it is strongly affected by hydrogen embrittlement. In general, the ferritic phase consists of a BCC structure and it is considered that hydrogen embrittlement susceptibility is high because hydrogen diffuses easily [10, 15]. Figure 6(b) shows the stress-strain curves of austenitic stainless steel SUSF310 under each test environment. Table 4 shows the ultimate tensile strength and maximum strain obtained from the stress-strain curves. Under a constant potential of −0.5 V (vs. Ag/AgCl), the maximum strain is similar value to that in corrosive solution and it is 14.3% less than in air. The use of the solution in an SCC environment resulted in some degradation of mechanical properties, however, the effect of hydrogen embrittlement was not as significant. In general, the austenitic phase consists of an FCC structure and it is considered that hydrogen embrittlement susceptibility is low because hydrogen diffuses hardly. However, it is known that metastable austenitic stainless steels such as SUS304 are susceptible to hydrogen embrittlement due to the formation of strain-induced martensite when plastic deformation is applied [10, 15, 24–27]. The SUSF310 used in this study contains a relatively large amount of Ni which is an austenitic phase forming element, at 20.0 mass%. Therefore, it is thought that strain-induced martensite was hardly formed and was not affected by hydrogen embrittlement.

Fig. 6

Stress-Strain curves of ferritic stainless steel SUS445J1 and austenitic stainless steel SUSF310 by SSRT under each test environment at 298 K. (online color)

Table 4 Ultimate tensile strength and maximum strain of ferritic stainless steel SUS445J1 and austenitic stainless steel SUSF310 derived from Stress-Strain curves.

As a result of the SSRT while cathodic charging was applied to each single-phase stainless steel, it was found that the ferritic phase has high hydrogen embrittlement susceptibility and the austenitic phase has low hydrogen embrittlement susceptibility. This suggests that the hydrogen embrittlement in F55 with area ratio of σ phase of 0% also occurs in mainly ferritic phase.

3.3 Fracture surface observation

After SSRT, the relationship between the test environment, metallographic structure and fracture morphology was observed by SEM. Figure 7(a) shows the fracture surface of super duplex stainless steel F55 with area ratio of σ phase of 0%. In air, dimples were observed on the fracture surface and this indicates that the fracture morphology was ductile fracture. In corrosive solution, dimples were not observed and cracks appeared. A brittle fracture surface was shown due to the influence of SCC. Under a constant potential of −0.5 V (vs. Ag/AgCl), dimples were not observed, and the fracture surface was brittle due to the effect of hydrogen embrittlement. Figure 7(b)–(e) shows the fracture surface of super duplex stainless steel F55 with area ratio of σ phase of 8.7%, 14.4%, 20.9% and 35.2%. In contrast to the specimen with area ratio of σ phase of 0%, dimples were not observed in the specimen with σ phase precipitation even in air and the fracture morphology was brittle fracture under all test environments. It is considered that the brittle fracture was caused by preferential cracking within the σ phase grains or at the interfaces adjacent to the sigma phase because the σ phase has hard and brittle properties. The specimens with σ phase precipitation showed similar brittle fracture surfaces under all test conditions, and no significant effects of SCC or hydrogen embrittlement on the fracture surfaces were observed.

Fig. 7

Fracture surface observation of super duplex stainless steel F55 (area ratio of σ phase: 0%, 8.7%, 14.4%, 20.9%, 35.2%) by SEM.

Figure 8(a) shows the fracture surface of ferritic stainless steel SUS445J1. In air and in corrosive solution, dimples were observed on the fracture surface and this indicates that the fracture morphology was ductile fracture. As with F55, under a constant potential of −0.5 V (vs. Ag/AgCl), a brittle fracture surface was observed. Figure 8(b) shows the fracture surface of austenitic stainless steel SUSF310. Under all test environments, dimples were observed and it can be confirmed that it has a certain degree of ductility even under the hydrogen embrittlement environment. As a result of SSRT and fracture surface observation of each single-phase stainless steel, it is unlikely that the austenitic phase causes hydrogen embrittlement. The hydrogen embrittlement in F55 with area ratio of σ phase of 0% is considered to be mainly caused by ferritic phase.

Fig. 8

Fracture surface observation of ferritic stainless steel SUS445J1 and austenitic stainless steel SUSF310 by SEM.

3.4 Surface observation

As a result of fracture surface observation, the fracture morphology of super duplex stainless steel F55 under each test environment was clarified. However, it is not clear which phase and interface affect the embrittlement of the specimen. Therefore, the surfaces of the specimens after SSRT under each test environment was observed to investigate the phase and interface through which the crack passed. Figure 9 shows the surface of super duplex stainless steel F55 with area ratio of σ phase of 0%, 8.7% and 35.2% after SSRT in air. In the case of area ratio of σ phase of 0%, no cracks were observed. In the case of area ratio of σ phase of 8.7% and 35.2%, many σ phase transgranular cracks were observed. Figure 10 shows the surface of super duplex stainless steel F55 with area ratio of σ phase of 0%, 8.7% and 35.2% after SSRT in corrosive solution. In the case of area ratio of σ phase of 0%, it can be seen that part of the ferritic phase is dissolved. Almost no cracks were observed in the specimens with precipitated σ phase. Based on the SSRT results, the mechanical properties deteriorate in corrosive solution, which is thought to be affected by SCC. However, the effect could not be confirmed in this surface observation. Figure 11 shows optical micrograph in a cross section of super duplex stainless steel F55 (area ratio of σ phase: 8.7%) after being immersed in a 2.5 mol·dm⁻³ H₂SO₄ + 0.2 mol·dm⁻³ NaCl solution for 518.4 ks. The ferritic phase around the σ phase is preferentially dissolved, and the σ phase with high Cr and Mo concentrations remains. Therefore, it is expected that embrittlement due to cracking of the APC-SCC around the sigma phase and the sigma phase itself is dominant in the SSRT in corrosive solution. However, in this experiment, SSRT was performed for a very long time, so preferential dissolution around the σ phase progressed, and the σ phase itself fell off from the surface. Therefore, no cracks were observed. Figure 12 shows the surface of super duplex stainless steel F55 with area ratio of σ phase of 0%, 8.7% and 35.2% after SSRT under a constant potential of −0.5 V (vs. Ag/AgCl). In the case of area ratio of σ phase of 0%, ferritic phase transgranular cracks and interface cracks between ferritic phases were observed. In the case of area ratio of σ phase of 8.7%, many ferritic phase transgranular cracks were observed, and almost no cracks were observed in the σ phase. In the case of area ratio of σ phase of 35.2%, many σ phase transgranular cracks were observed. Under a constant potential of −0.5 V (vs. Ag/AgCl), many cracks were observed, even though the same solution was used as in the corrosion solution. In the cathodic region, preferential dissolution around the σ phase does not occur because the anodic reaction is suppressed, and the σ phase remains. Therefore, under a constant potential of −0.5 V (vs. Ag/AgCl), it is thought that σ phase embrittlement occurs in addition to hydrogen embrittlement, susceptibility to cracking increased. On the other hand, in corrosive solution, the σ phase that would normally form as cracks fell off the surface, and susceptibility to cracking decreased. Therefore, it showed better mechanical properties than under a constant potential of −0.5 V (vs. Ag/AgCl).

Fig. 9

Optical micrographs in the surface of super duplex stainless steel F55 (area ratio of σ phase: 0%, 8.7%, 35.2%) after SSRT in air. The micrographs are a vertical field of view of the flat portion of the gauge. (online color)

Fig. 10

Optical micrographs in the surface of super duplex stainless steel F55 (area ratio of σ phase: 0%, 8.7%, 35.2%) after SSRT in corrosive solution. The micrographs are a vertical field of view of the flat portion of the gauge. (online color)

Fig. 11

Optical micrograph in a cross section of super duplex stainless steel F55 (area ratio of σ phase: 8.7%) after being immersed in a corrosive solution for 518.4 ks. (online color)

Fig. 12

Optical micrographs in the surface of super duplex stainless steel F55 (area ratio of σ phase: 0%, 8.7%, 35.2%) after SSRT under a constant potential of −0.5 V (vs. Ag/AgCl). The micrographs are a vertical field of view of the flat portion of the gauge. (online color)

Table 5–9 shows the results of calculating the rate of cracks passing through each phase and interface in the surface of the specimens after SSRT in air and under a constant potential of −0.5 V (vs. Ag/AgCl). The total length of the cracks measured was at least 100 µm, and it was statistically determined which phases and interfaces were susceptible to cracking. In the case of area ratio of σ phase of 0%, no cracks were observed in air. Under a constant potential of −0.5 V (vs. Ag/AgCl), ferritic phase transgranular cracks and interface cracks between ferritic phases were observed at similar rates. All of these cracks stopped when they reached the austenitic phase. No cracks were observed in the austenitic phase grains and austenitic phase interfaces. X.Z. Liang et al. reported that tensile tests of duplex stainless steel with cathodic charging showed that most of the internal cracks propagated in the ferritic phase and were blocked by the austenitic phase. In other words, the ferritic phase was assumed to be the preferential path for cracking due to hydrogen embrittlement [15, 16, 18].

Table 5 Rate of cracks passing through each phase and interface in super duplex stainless steel F55 (area ratio of σ phase: 0%) after SSRT.

Table 6 Rate of cracks passing through each phase and interface in super duplex stainless steel F55 (area ratio of σ phase: 8.7%) after SSRT.

Table 7 Rate of cracks passing through each phase and interface in super duplex stainless steel F55 (area ratio of σ phase: 14.4%) after SSRT.

Table 8 Rate of cracks passing through each phase and interface in super duplex stainless steel F55 (area ratio of σ phase: 20.9%) after SSRT.

Table 9 Rate of cracks passing through each phase and interface in super duplex stainless steel F55 (area ratio of σ phase: 35.2%) after SSRT.

Therefore, in the case of area ratio of σ phase of 0%, hydrogen embrittlement is considered to occur mainly within the ferritic phase grains and at the interfaces between ferritic phases. This is because the hydrogen introduced by cathodic charging diffuses mainly within the ferritic phase grains with BCC structure and interfaces between ferritic phases. In contrast, the austenitic phase with the FCC structure is highly resistant to hydrogen embrittlement, and is responsible for stopping cracking. Figure 13 shows the rate of cracks that passed through the σ phase among the cracks observed in air and at a constant potential of −0.5 V (vs. Ag/AgCl). In the case of area ratio of σ phase of 8.7%, σ phase transgranular cracks occupy the majority in air and σ phase embrittlement is dominant. Under a constant potential of −0.5 V (vs. Ag/AgCl), the rate of σ phase transgranular cracks decreased significantly compared to that in air. Moreover, ferritic phase transgranular cracks which are thought to be caused by hydrogen embrittlement occupy the majority. In addition, austenitic phase transgranular cracks and σ phase transgranular cracks were slightly observed. However, since the rate of these is less than 10%, it is thought that they simply passed through ferritic phase during the progress of cracks. It is difficult to think that these cracks greatly affect the embrittlement of the specimens. Therefore, in the case of a relatively small amounts of area ratio of σ phase of 8.7%, the specimen is embrittled by a combination of σ phase embrittlement and hydrogen embrittlement, but the effect of hydrogen embrittlement is considered to be dominant. A similar tendency was observed with area ratio of σ phase of 14.4% and 20.9%. In air, σ phase transgranular cracks occupy the majority and σ phase embrittlement is dominant. Under a constant potential of −0.5 V (vs. Ag/AgCl), unlike the case of area ratio of σ phase of 8.7%, σ phase transgranular cracks occupy the majority. In addition, the rate of ferritic phase transgranular cracks tends to increase, confirming the effect of hydrogen embrittlement. Although, in the case of a relatively small amounts of area ratio of σ phase of 8.7%, the specimen is embrittled by a combination of σ phase embrittlement and hydrogen embrittlement, the effect of hydrogen embrittlement is considered to be dominant. Although, in the case of area ratio of σ phase of 14.4% and 20.9%, the specimen is embrittled by a combination of σ phase embrittlement and hydrogen embrittlement, the effect of σ phase embrittlement is considered to be dominant. In the case of area ratio of σ phase of 35.2%, regardless of test environment, σ phase transgranular cracks occupy the majority and σ phase embrittlement is dominant. Considering the fact that the maximum strain obtained from SSRT did not change in air and under a constant potential of −0.5 V (vs. Ag/AgCl), it can be said the influence of hydrogen embrittlement was hidden in the embrittlement due to a large amount of precipitated σ phase. Moreover, under a constant potential of −0.5 V (vs. Ag/AgCl), the rate of secondary austenitic phase transgranular cracks increased compared to that in air. This suggests that the secondary austenitic phase precipitated by reheating the specimen may be susceptible to influence of hydrogen embrittlement. However, this tendency was not confirmed in other specimens. In the specimens with a large amount of σ phase precipitated, a large amount of secondary austenitic phase was also precipitated. Therefore, it is also possible that the σ phase transgranular cracks simply passed through the nearby secondary austenitic phase.

Fig. 13

Rate of cracks passing through σ phase on the surface of super duplex stainless steel F55 after SSRT in air and under a constant potential of −0.5 V (vs. Ag/AgCl). (online color)

According to these results, hydrogen embrittlement is dominant in the case of a relatively small amounts of σ phase area ratio of 8.7%, while σ phase embrittlement is dominant in the case of σ phase area ratio of 14.4% and 20.9% and 35.2%. Therefore, it cannot be said that the σ phase promotes hydrogen embrittlement.

Surface observation of the specimens show that the σ phase precipitates mainly at the interface between the ferritic phases and austenitic phases in the case of a relatively small amounts of σ phase area ratio of 8.7%. On the other hand, in the case of σ phase area ratio of 14.4% or higher, a large amount of σ phase was precipitated not only at the interface but also within the ferritic phase. Under hydrogen embrittlement, transgranular cracks occurs preferentially in the ferritic phase. However, in the case of a σ phase area ratio of 14.4% or higher, a large rate of the cracks pass through the σ phase during the cracking process because the σ phase is also precipitated within the ferritic phase grains. It is considered that cracks preferentially pass through the σ phase because the σ phase is more susceptible to cracking than the ferritic phase.

4. Conclusions

The purpose of this paper is to clarify the effect of σ phase on the SCC susceptibility and hydrogen embrittlement susceptibility of super duplex stainless steel F55. SSRT was performed in air, in corrosive solution and under cathodic charging using five types of specimens with different area ratio of σ phase. Moreover, we observed the fracture surface and surface of the specimens after SSRT. The findings obtained are shown below.

1. (1)    As a result of SSRT in corrosive solution, the specimens with precipitated σ phase tended to have a more pronounced decrease in ductility. Therefore, it can be thought that the presence of σ phase promotes SCC. This is because the ferritic phase surrounding the σ phase dissolves preferentially, increasing APC-SCC susceptibility.
2. (2)    As a result of SSRT under cathodic charging, the ultimate tensile strength and maximum strain showed constant low values regardless of the area ratio of σ phase. Moreover, comparing the ultimate tensile strength and maximum strain in SSRT under cathodic charging with those in air, there was a tendency for the reduction rate to decrease as the area ratio of σ phase increased. Therefore, σ phase did not promote hydrogen embrittlement.
3. (3)    Under cathodic charging, the specimens with precipitated σ phase are embrittled by a combination of σ phase embrittlement and hydrogen embrittlement. In the case of a relatively small amounts of area ratio of σ phase of 8.7%, hydrogen embrittlement is dominant. In the case of a relatively large amounts of area ratio of σ phase of 14.4%, 20.9% and 35.2%, σ phase embrittlement is dominant.
4. (4)    It showed better mechanical properties in corrosive solution than under cathodic charging. σ phase drops off from the surface due to preferential dissolution around the σ phase. In this way, the cracking susceptibility was reduced because the starting point for cracking was eliminated. On the other hand, under cathodic charging, the anodic reaction is suppressed and the σ phase remains because the potential is maintained in the cathodic region. Therefore, it is thought that the combination of hydrogen embrittlement and σ phase embrittlement increased susceptibility to cracking and significantly decreased mechanical properties.

Acknowledgement

We are grateful to the Pacific Steel Mfg. Co., Ltd provided the specimens for this study. This study was financially supported by the Grant-in-Aid for Scientific Research (C) (No. 22K04774).

REFERENCES

• 1)   T.  Zakroczymski,  A.  Glowacka and  W.  Swiatnicki: Effect of hydrogen concentration on the embrittlement of a duplex stainless steel, Corros. Sci. 47 (2005) 1403–1414. doi:10.1016/j.corsci.2004.07.036
• 2)   G.  Ogita,  K.  Matsumoto,  M.  Mochizuki,  Y.  Mikami and  K.  Ito: Evaluation of Hydrogen-induced Cracking Behavior in Duplex Stainless Steel by Numerical Simulation of Stress and Diffusible Hydrogen Distribution at the Microstructural Scale, ISIJ Int. 61 (2021) 1135–1142. doi:10.2355/isijinternational.ISIJINT-2020-399
• 3)   A.  Yousefi and  G.  Itoh: Tensile Properties of an Electrolytically Hydrogen Charged Duplex Stainless Steel Affected by Strain Rate, ISIJ Int. 58 (2018) 561–565. doi:10.2355/isijinternational.ISIJINT-2017-631
• 4)   P.  Craidy,  L.  Briottet and  D.  Santos: Hydrogen-Microstructure-Mechanical properties interactions in super duplex stainless steel components, Int. J. Hydrogen Energ. 40 (2015) 17084–17090. doi:10.1016/j.ijhydene.2015.06.065
• 5)   W.  Wu,  Z.  Liu,  S.  Hu,  X.  Li and  C.  Du: Effect of pH and hydrogen on the stress corrosion cracking behavior of duplex stainless steel in marine atmosphere environment, Ocean Eng. 146 (2017) 311–323. doi:10.1016/j.oceaneng.2017.10.002
• 6)   E.M.  Cojocaru,  D.  Raducanu,  A.  Nocivin and  V.D.  Cojocaru: Influence of ageing treatment temperature and duration on r-phase precipitation and mechanical properties of UNS S32750 SDSS alloy, J. Adv. Res. 30 (2021) 53–61. doi:10.1016/j.jare.2020.11.005
• 7)   N.  Lopez,  M.  Cid and  M.  Puiggali: Influence of σ-phase on mechanical properties and corrosion resistance of duplex stainless steels, Corros. Sci. 41 (1999) 1615–1631. doi:10.1016/S0010-938X(99)00009-8
• 8)   K.  Nakade and  T.  Kuroda: Precipitation Mechanism of Sigma Phase in Super Duplex Stainless Steels, J. High Temp. Soc. 33 (2007) 95–100. doi:10.7791/jhts.33.95
• 9)   K.  Nagae,  K.  Ejiri,  J.  Hirabayashi,  Y.  Yamamoto,  M.  Hatakeyama and  S.  Sunada: Effect of Sigma Phase on Corrosion Behavior of Duplex Stainless Steel, Mater. Trans. 63 (2022) 726–729. doi:10.2320/matertrans.MT-MA2022002
• 10)   M.  Okayasu and  T.  Fujiwara: Effects of microstructural characteristics on the hydrogen embrittlement characteristics of austenitic, ferritic, and γ-α duplex stainless steels, Mater. Sci. Eng. A 807 (2021) 140851. doi:10.1016/j.msea.2021.140851
• 11)   Y.  Shimada,  M.  Hatakeyama,  J.  Hirabayashi,  Y.  Yamamoto,  K.  Sato and  S.  Sunada: Effects of Sigma Phase on the Embrittlement Behavior of Super Duplex Stainless Steel F55 under SSRT with a Cathodic Charge, Mater. Trans. 60 (2019) 2378–2386. doi:10.2320/matertrans.MT-MAW2019006
• 12)   A.  Świerczyńska,  D.  Fydrych,  M.  Landowski,  G.  Rogalski and  J.  Łabanowski: Hydrogen embrittlement of X2CrNiMoCuN25-6-3 super duplex stainless steel welded joints under cathodic protection, Constr. Build. Mater. 238 (2020) 117697. doi:10.1016/j.conbuildmat.2019.117697
• 13)   T.  Wang,  X.  Fang,  W.  Lv,  H.  Zhang,  Y.  Luo,  L.  Zheng and  W.  Liang: Effect of hydrogen on the mechanical properties and fracture modes of annealed 430 ferritic stainless steel, Mater. Sci. Eng. A 832 (2022) 142491. doi:10.1016/j.msea.2021.142491
• 14)   A.  Komatsu,  M.  Fujinami,  M.  Hatano,  K.  Matsumoto,  M.  Sugeoi and  L.  Chiari: Straining-temperature dependence of vacancy behavior in hydrogen-charged austenitic stainless steel 316L, Int. J. Hydrogen Energ. 46 (2021) 6960–6969. doi:10.1016/j.ijhydene.2020.11.148
• 15)   X.Z.  Liang,  G.-H.  Zhao,  M.F.  Dodge,  T.L.  Lee,  H.B.  Dong and  P.E.J.  Rivera-Diaz-Castillo: Hydrogen embrittlement in super duplex stainless steels, Materialia 9 (2020) 100524. doi:10.1016/j.mtla.2019.100524
• 16)   S.-L.  Chou and  W.-T.  Tsai: Hydrogen embrittlement of duplex stainless steel in concentrated sodium chloride solution, Mater. Chem. Phys. 60 (1999) 137–142. doi:10.1016/S0254-0584(99)00077-2
• 17)   T.  Chida,  Y.  Hagihara,  E.  Akiyama,  K.  Iwanaga,  S.  Takagi,  H.  Ohishi,  M.  Hayakawa,  D.  Hirakami and  T.  Tarui: Comparison of Constant Load, SSRT and CSRT Methods for Hydrogen Embrittlement Evaluation Using Round Bar Specimens of High Strength Steels, Tetsu-to-Hagané 100 (2014) 1298–1305. doi:10.2355/tetsutohagane.100.1298
• 18)   A.M.  Elhoud,  N.C.  Renton and  W.F.  Deans: Hydrogen embrittlement of super duplex stainless steel in acid solution, Int. J. Hydrogen Energ. 35 (2010) 6455–6464. doi:10.1016/j.ijhydene.2010.03.056
• 19)   S.S.M.  Tavares,  R.T.  Batista,  R.V.  Landim,  J.A.C.  Velasco and  L.F.  Senna: Investigation of the effect of low temperature aging on the mechanical properties and susceptibility to sulfide stress corrosion cracking of 22%Cr duplex stainless steel, Eng. Fail. Anal. 113 (2020) 104553. doi:10.1016/j.engfailanal.2020.104553
• 20)   J.  Rajaguru and  N.  Arunachalam: Effect of machined surface integrity on the stress corrosion cracking behavior of super duplex stainless steel, Eng. Fail. Anal. 125 (2021) 105411. doi:10.1016/j.engfailanal.2021.105411
• 21)   J.-S.  Lee,  K.  Fushimi,  T.  Nakanishi,  Y.  Hasegawa and  Y.-S.  Park: Corrosion behaviour of ferrite and austenite phases on super duplex stainless steel in a modified green-death solution, Corros. Sci. 89 (2014) 111–117. doi:10.1016/j.corsci.2014.08.014
• 22)   S.  Sunada,  M.  Kariba,  K.  Majima and  K.  Sugimoto: Influences of Concentrations of H₂SO₄ and NaCl on Stress Corrosion Cracking of SUS304 Stainless Steel in H₂SO₄ - NaCl Aqueous Solutions, Mater. Trans. 47 (2006) 364–370. doi:10.2320/matertrans.47.364
• 23)   K.  Nakade and  T.  Kuroda: Interaction between Sigma Phase and Hydrogen in Duplex Stainless Steel Weld Metal, Zairyo 54 (2005) 239–244. doi:10.2472/jsms.54.239
• 24)   L.R.  Queiroga,  G.F.  Marcolino,  M.  Santos,  G.  Rodrigues,  C.  Eduardo dos Santos and  P.  Brito: Influence of machining parameters on surface roughness and susceptibility to hydrogen embrittlement of austenitic stainless steels, Int. J. Hydrogen Energ. 44 (2019) 29027–29033. doi:10.1016/j.ijhydene.2019.09.139
• 25)   A.  Ueno and  G.  Benjamin: Effect of high-pressure H₂ gas on tensile and fatigue properties of stainless steel SUS316L by means of the internal high-pressure H₂ gas method, Procedia Struct. Integr. 19 (2019) 494–503. doi:10.1016/j.prostr.2019.12.053
• 26)   A.  Macadre,  T.  Masumura,  R.  Manabe,  T.  Tsuchiyama and  S.  Takaki: Effect of nitrogen-addition on the absorption and diffusivity of hydrogen in a stable austenitic stainless steel, Int. J. Hydrogen Energ. 44 (2019) 1263–1271. doi:10.1016/j.ijhydene.2018.11.031
• 27)   X.  Li,  M.  Gao,  H.  Li,  W.  Xing,  L.  Zhang,  L.  Shu,  X.  Zhao,  Y.  Ma and  K.  Liu: Effect of residual hydrogen content on the tensile properties and crack propagation behavior of a type 316 stainless steel, Int. J. Hydrogen Energ. 44 (2019) 25054–25063. doi:10.1016/j.ijhydene.2019.07.131